Mike White Hotels - score:* 2.9329195*
Think Twice - score:* 2.756144*
White House Hotel - score: 2.2049153
Nice Time Hotel - score: 2.065069
De Wise Hotel Odo Ona - score: 1.9627647
Under The Wave - score: 1.6577798
*
*
The result is fine, but i need the ordering to be better, that is, I need
'De Wise Hotel' to be matched first with a higher score before other fuzzy
calculated results.
I'm new to elastic search, so your help would be greatly appreciated.
I expect this is down to a core Lucene issue in that the default FuzzyQuery
class has a bias towards rare terms rather than good matches. I addressed
this problem with the "FuzzyLikeThis" alternative Query class.
Elasticsearch docs for this alternative are here
I haven't tried this in ES but I have generally found FLT provides saner
ranking of results.
Cheers
Mark
On Wednesday, September 11, 2013 9:00:03 AM UTC+1, Makanju Dami Perfect
wrote:
Hello,
I'm trying to create a correction search for one of my projects. And i'm
currently using fuzzy search, but the scoring of the results are not
helpful.
Mike White Hotels - score:* 2.9329195*
Think Twice - score:* 2.756144*
White House Hotel - score: 2.2049153
Nice Time Hotel - score: 2.065069
De Wise Hotel Odo Ona - score: 1.9627647
Under The Wave - score: 1.6577798
*
*
The result is fine, but i need the ordering to be better, that is, I need
'De Wise Hotel' to be matched first with a higher score before other fuzzy
calculated results.
I'm new to Elasticsearch, so your help would be greatly appreciated.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.